yan. I/, 1889] 



NATURE 



273 



more or less combined silica, which may reach to upwards 

 of 65 per cent, of the mass : these are the acid or light lavas. 

 Thence we pass, by various gradations, to the basic or 

 dense lavas, in which the proportion of silica, gradually 

 diminishing, does not reach more than 55 or even 45 per 

 rent. This silica does not e.xist in a free state in modern 

 lavas, but is combined, in the form of silicates, with alumina, 

 iron, lime, magnesia, potash, and soda. 



In the slags of metallurgical works we find products 

 which present close analogy to those of volcanoes, both in 

 composition and in mode of formation. These artificial 

 scoriae are, like lavas, formed of silicates ; and another 

 point of resemblance between them lies in the fact that 

 we may regard both as the scum of a metallic nucleus, of 

 which they form the upper zones. The differences in com- 

 position result from the fact that they are derived from 

 2ones of greater or less depth. 



Our knowledge of eruptive rocks came to be enriched in 

 an unexpected manner by the application of the micro- 

 scope to liihology. We need not here recall the almost 

 marvellous results obtained by this method of investiga- 

 tion, inaugurated by H. C. Soiby ; but we may say, in a 

 word, that the microscopic analysis of rocks has changed 

 the face of petrography. Let us confine our attention to 

 some of the conceptions relating to modern volcanic rocks, 

 as revealed by these new methods — methods which, in 

 delicacy, in certainty, and in elegance, are unsurpassed in 

 any other branch of natural science. Not only have they 

 enabled us to verify and control hypotheses, but they have 

 led to the remarkable discoveries' to which I am about to 

 refer. 



The eye, assisted even by the most powerful lenses^ could 

 recognize in lavas only those minerals which appeared in 

 rather large crystals ; chemical analysis generally gave 

 merely the composition of the total rock, and its rninera- 

 logical composition was only suspected. The intimate 

 texture of the rock remained impenetrable ; it was im- 

 possible to determine with certainty the order in which the 

 constituents of the molten mass had solidified ; neither 

 could we trace the various states through which the crystals 

 had passed — their germs, primordial forms, and skeletons 

 —or the aspect of the rock at different stages of its 

 development. 



Let us now apply the microscope to the examination 

 of a thin slice of lava, rendered transparent by polishing. 

 The lavas, as we have said, may be compared to vitreous 

 masses ; but whilst in our artificial glasses we seek to 

 obtain a pellucid and homogeneous product, the liquefied 

 matter of volcanoes, when it flows forth, already contains 

 certain differentiated products. The glass which contains 

 these bodies may be regarded as the residue of the 

 crystallization, whence the numerous crystalline indivi- 

 duals have extracted their constituent elements. In the 

 black, brilliant, volcanic glasses, apparently opaque and 

 destitute of crystallization, the microscope discovers a 

 world of mineral forms. It shows us their various states of 

 growth, and the arrest of their development consequent 

 on the more or less rapid consolidation of the mass. It 

 is especially in those rocks which, like obsidian, have pre- 

 served almost wholly their vitreous character, and are 

 homogeneous to the naked eye, that we find the rudimen- 

 tary crystals of curious form, representing the first step in 

 the passage of the amorphous matter to the crystalline 

 condition. Owing to the rapidity with which the vitreous 

 paste consolidated, the crystals were unable to grow, and 

 their development was sharply arrested. Hence the origin 

 of these embryonic crystals which abound in natural 

 glasses, and which we designate as crystallites. Ana- 

 logous crystallites are produced in blast-furnace slags, 

 which have close relations ta the matter of lavas. Their 

 conimon origin is betrayed by certain family likenesses 

 which the microscope reveals. The slags, examined in 

 thin sections, exhibit rudimentary crystalline forms, 

 similar to the crystallites of volcanic glasses. 



But usually the crystals have not remained in this em- 

 bryonic state. If the lava has not been too rapidly 

 cooled, the molecular movements are retained, even in ;i 

 semi-liquid mass, and the paste develops crystals of 

 minute dimensions, called microlites. These microscopic 

 crystals are formed in the heart of the vitreous magma 

 during its slow consolidation. Notwithstanding their 

 infinite minuteness, these small polyhedra exhibit with 

 marvellous e.xactitude all their specific characteristics, 

 such as we are familiar with in much larger crystals, and 

 which we should not expect to find in lavas. They often 

 form, by their interlacement, a beautiful network in the 

 paste, and give to the rock in which they are developed a 

 microlitic structure. 



The dimensions of these microlites, invariably micro- 

 scopic, and their arrangement, prove that they may be 

 referred to a period of disturbance ; that they were formed, 

 indeed, at a time when the lava, though still in motion, 

 was solidifying. They separated from the magma during 

 the very act of outflow or eruption. 



Besides these microscopic crystals and these groups of 

 crystallites, which belong to the last stage of consolida- 

 tion, the lava contains also a supply of larger crystals, 

 more fully developed, and in many cases recognizable by 

 the naked eye. These have been formed under calmer 

 conditions, analogous to those presented by a tranquil 

 fluid in which crystallization is proceeding slowly. They 

 were formed in the molten magma when it was still in- 

 closed in the subterranean reservoirs. This slow growth 

 is clearly proved by the formation of the crystals in con- 

 centric zones and by their size. These large crystals, 

 existing ready formed in the lava at the time of its 

 eruption, are surrounded by microlites or by a vitreous 

 mass. It was after their slow development in the magma, 

 during an intra-telluric period, that the mass in which 

 they floated was upraised. The period of calm was 

 succeeded by one of agitation, and the lava in its violent 

 ejection carried forth the crystals, breaking them, corrod- 

 ing them, and partially fusing them. The microscope offers 

 distinct evidence of these phenomena. We see the large 

 crystals dislocated and their fragments dispersed, their 

 edges rounded and eroded, and their substance invaded 

 and penetrated by the paste. 



While the physical and chemical agencies brought into 

 play by the movement of the lava thus attack the ancient 

 crystals to the verge of demolition, the microlites are in 

 course of formation. This vitreous matter, in which the 

 large crystals float, solidifies as a mass of microscopic 

 individuals. The latter are therefore related to a second 

 phase of crystallization : they are developed in a moving 

 viscous magma, and their further growth is arrested by 

 the rapid cooling which induces solidification eft masse. 



The fluidal arrangement of the microlites distinctly 

 shows, too, that the crystalline action was contempora- 

 neous with the movement of the lava-flow. Indeed, we 

 see in microscopic preparations that the microlites are 

 accumulated around the large sections of crystals, forming 

 wavy trains and presenting the arrangement which micro- 

 gra^htrs de%\gx\2i\.e. xisjliiiilal structure. It is marked by 

 the orientation of these infinitely small acicular crystals. 

 When these streams of microlites meet the large em- 

 bedded crystals, they sweep round them, crowding into 

 the spaces between the large sections, accommodatin.:^ 

 their flow to these outlines, and preserving for us the 

 last movement of the mass at the very moment of 

 solidification. 



The microscope therefore proves that crystallization in 

 lavas belongs to two periods : the first, anterior to the 

 eruption, during which the large crystals already found 

 are suspended in a mass that we may regard as entirely 

 vitreous ; and the second period, when the microlites and 

 embryonic crystalline forms are separated, dating frorr. 

 the ejection or outflow, and contemporaneous with the 

 solidification of the rock. 



